import pandas as pd
import numpy as np
print(pd.__version__)

rng = np.random.default_rng(seed=44)

expense_categories = {
    'Rent': (1400, 1400),
    'Groceries': (60, 160),
    'Utilities': (80, 220),
    'Dining Out': (15, 90),
    'Entertainment': (10, 70),
    'Transport': (20, 100)
}
print(list(expense_categories.keys()))

rows = []
months = pd.date_range('2026-01-01', periods=6, freq='MS')

for month_start in months:
    rows.append([month_start + pd.Timedelta(days=0), 'Salary Deposit', 'Income', round(float(rng.uniform(3800, 4200)), 2)])
    rows.append([month_start + pd.Timedelta(days=14), 'Salary Deposit', 'Income', round(float(rng.uniform(3800, 4200)), 2)])
    rows.append([month_start + pd.Timedelta(days=1), 'Monthly Rent', 'Rent', -1400.0])
    n_extra = int(rng.integers(10, 18))
    for _ in range(n_extra):
        category = rng.choice([c for c in expense_categories if c != 'Rent'])
        low, high = expense_categories[category]
        day_offset = int(rng.integers(0, 28))
        amount = -round(float(rng.uniform(low, high)), 2)
        rows.append([month_start + pd.Timedelta(days=day_offset), f'{category} purchase', category, amount])

transactions = pd.DataFrame(rows, columns=['date', 'description', 'category', 'amount'])
print(transactions.shape)

transactions.to_csv('personal_transactions.csv', index=False)
transactions = pd.read_csv('personal_transactions.csv', parse_dates=['date'])
transactions = transactions.sort_values('date').reset_index(drop=True)
transactions.head()

print(transactions.shape)
transactions.info()

transactions['category'].value_counts()

transactions['type'] = np.where(transactions['amount'] > 0, 'Income', 'Expense')
transactions[['description', 'category', 'amount', 'type']].head()

total_income = transactions.loc[transactions['type'] == 'Income', 'amount'].sum()
total_expense = transactions.loc[transactions['type'] == 'Expense', 'amount'].sum()
print(f'Total income: ${total_income:,.2f}')
print(f'Total expenses: ${total_expense:,.2f}')
print(f'Net over 6 months: ${total_income + total_expense:,.2f}')

transactions['month'] = transactions['date'].dt.to_period('M')
monthly_summary = transactions.groupby(['month', 'type'])['amount'].sum().unstack(fill_value=0)
monthly_summary['net'] = monthly_summary['Income'] + monthly_summary['Expense']
monthly_summary.round(2)

monthly_summary['savings_rate_pct'] = (monthly_summary['net'] / monthly_summary['Income'] * 100).round(1)
monthly_summary[['Income', 'Expense', 'net', 'savings_rate_pct']]

expenses_only = transactions[transactions['type'] == 'Expense']
category_totals = expenses_only.groupby('category')['amount'].sum().abs().sort_values(ascending=False)
category_totals

pivot = pd.pivot_table(expenses_only, values='amount', index='month', columns='category', aggfunc='sum', fill_value=0).abs().round(2)
pivot

starting_balance = 2000
transactions['running_balance'] = starting_balance + transactions['amount'].cumsum()
transactions[['date', 'description', 'amount', 'running_balance']].tail()

import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

fig, ax = plt.subplots(figsize=(8, 5))
colors = ['seagreen' if v >= 0 else 'crimson' for v in monthly_summary['net']]
monthly_summary['net'].plot(kind='bar', ax=ax, color=colors)
ax.set_title('Net Income by Month')
ax.set_ylabel('Net ($)')
ax.axhline(0, color='black', linewidth=0.8)
plt.tight_layout()
plt.savefig('monthly_net.png', dpi=150)
plt.close(fig)
print('Saved monthly_net.png')

fig, ax = plt.subplots(figsize=(8, 5))
category_totals.plot(kind='bar', ax=ax, color='darkorange')
ax.set_title('Total Spending by Category (6 Months)')
ax.set_ylabel('Spending ($)')
plt.tight_layout()
plt.savefig('spending_by_category.png', dpi=150)
plt.close(fig)
print('Saved spending_by_category.png')